Omneky Launches Tool to Evaluate AI-Generated Ad Quality
Omneky has introduced TASTE BENCH, an evaluation suite that tests how well AI-generated image and video creatives meet brand briefs and assets. The system assesses brand fit and creative execution.
San Francisco, CA – AI growth platform Omneky has launched TASTE BENCH, an evaluation suite designed to test how effectively AI models translate creative briefs and brand assets into ready-to-run advertisements.
The new suite assesses the quality of image and video creatives, their suitability for a brand, and their readiness for deployment. Initial tests showed the GPT Image 2.5 Sunburst model achieved the highest "ready-to-run" rate at 54.2 percent, with an average quality score of 7.38 out of 10.
TASTE BENCH compares outputs from eight different AI image models using identical prompts and brand assets. Results provide detailed insights into creative quality, evaluation costs, and readiness, offering transparency into model performance.
Founded in 2018 and serving over 6,000 customers, Omneky utilizes AI to create and optimize ad campaigns. The development of TASTE BENCH reflects the company's commitment to ensuring high quality and effectiveness in AI-generated advertising material.